The AI Jobs Shock Is a Rebalancing, Not a Collapse

The AI jobs debate has the wrong headline. Mass unemployment is a long run argument nobody can prove yet; getting out-operated by a leaner competitor within 12 months is a risk you’re already carrying.

Key takeaways

The macro jobs debate is a distraction from an operational risk that’s already live
Companies that redesign roles around AI now will out-execute slower rivals fast
Headcount isn’t the metric that matters; output per person is
Waiting for certainty on the labour market means waiting too long to act
Board level conversations should focus on role design, not job security theatre

I’ve built and sold two companies, Obby and Baluu, and I now run FirstMotion alongside advising founders and leadership teams through VAQA. In every one of those seats, the thing that decided who won wasn’t headcount, it was how quickly a team turned new tools into output. This piece is about why I think the “will AI cause mass unemployment” debate is a sideshow, and what leaders should actually be doing with the next two quarters.

Why the macro debate misses the point

Every few weeks another think tank or economist publishes a forecast about AI and jobs, and the range of predictions is enormous. Some say millions of roles vanish within a decade; others say history shows technology creates as many jobs as it destroys. I genuinely don’t know who’s right, and honestly, neither do they.

Here’s what I do know. While that argument plays out in op-eds, individual companies are already restructuring how work gets done, and the ones doing it well are pulling ahead of the ones still debating it. That gap opens long before any macro data confirms a trend either way.

Waiting for consensus on a ten year question is a terrible strategy when your competitor is redesigning their operations this quarter.

The rebalancing that’s already happening

What I’m actually seeing across the businesses I advise isn’t collapse, it’s reallocation. Tasks that used to need a person now need a person plus a tool, or in some cases just a tool with a person checking it.

That’s a rebalancing of where human time goes, not a disappearance of human value. A few patterns show up again and again:

First line customer support increasingly triaged by AI, with people handling the escalations that need judgement
Content and research functions shifting from producing first drafts to editing and verifying them
Sales development roles narrowing toward qualification and relationship work, with prospecting largely automated
Finance and ops teams spending less time on reconciliation and more on decisions

None of that is mass unemployment. It’s a redistribution of effort toward the parts of a job that still need a human, and away from the parts that don’t.

Why the redesign has to happen now, not eventually

Here’s the part leaders underestimate. A company that hasn’t rebuilt its workflows around AI isn’t standing still while everyone else stands still too. It’s actively falling behind a competitor who has.

I’ve watched this pattern up close through FirstMotion’s work with B2B SaaS companies. A team of eight that’s redesigned its go-to-market motion around AI tooling can often outproduce a team of fifteen that hasn’t touched its processes.

That’s not a five year gap. In software and services markets, I’ve seen that kind of operational lead open up within two or three quarters. By the time the macro data on jobs finally settles the debate, the leaders who acted will already have compounded advantages in cost base, speed, and margin.

What “redesigning roles” actually means in practice

This isn’t about firing people and calling it AI transformation. It’s a genuinely different exercise, and most leadership teams haven’t done it properly yet.

It starts with mapping what a role actually does, task by task, rather than what the job title implies. Most job descriptions were written for a world where every task required a human doing it manually from start to finish. That’s no longer true for a large share of knowledge work.

Once you have that map, the question becomes which tasks can move to AI assisted or AI led execution, and which genuinely need a person’s judgement, relationships, or accountability. I’d say most leadership teams I meet through VAQA’s board advisory work haven’t done this exercise rigorously. They’ve bought tools and hoped for the best, rather than redesigning the operating model the tools sit inside.

The competitive risk nobody’s pricing in properly

Boards spend a lot of time on downside protection: compliance, security, reputational risk. Fewer spend equivalent time on the risk of simply being slower than a leaner competitor.

That’s the risk I think is underpriced right now. It doesn’t show up in a risk register because it’s not a single event, it’s a gradual erosion of margin and speed that compounds quietly until a competitor’s pricing or delivery speed becomes impossible to match.

Organisations like OpenAI and Anthropic keep shipping capability into the workplace faster than most companies can absorb it. The absorption gap, not the tool itself, is where competitive advantage is being won or lost right now.

Rebalancing, not collapse, and that’s the harder problem

I’d rather face a straightforward collapse than a quiet rebalancing, honestly. A collapse forces everyone to react. A rebalancing lets you convince yourself you’ve got more time than you actually do.

The uncomfortable truth is that the companies winning this transition aren’t waiting for certainty about the labour market. They’re redesigning roles around what AI can already do reliably, and adjusting again every few months as the tools improve.

That’s a harder discipline than either “AI will take all the jobs” or “AI is overhyped, nothing will change.” It requires ongoing, unglamorous operational work, which is exactly why most organisations avoid it until a competitor forces their hand.

Where to start if you haven’t

If you’re leading a team right now and haven’t done a proper role by role audit against what AI tools can already do, that’s the highest leverage use of your time this quarter. Not another all hands about “our AI strategy,” an actual task level map of where effort goes and where it should go instead.

I help founders and leadership teams work through exactly this under VAQA’s Growth and Operational Efficiency pillars, alongside the go to market and finance questions that usually come with it. If you want a second pair of eyes on where your organisation is exposed, get in touch and let’s talk through it. You can also see what I advise on if you want the full picture first.

Frequently Asked Questions

Is AI actually going to cause mass unemployment?

Nobody can say for certain, and I’d be sceptical of anyone who claims otherwise with confidence. What’s much more certain is that individual companies redesigning their operations around AI now will outperform those waiting for macro clarity, regardless of how the long run numbers eventually land.

What does “redesigning roles around AI” mean in practice?

It means mapping what a job actually involves task by task, rather than relying on the job title, and deciding which tasks move to AI assisted execution versus which still need human judgement. Most teams skip this and just add tools on top of unchanged workflows, which limits the benefit considerably.

How quickly can a competitor really gain an edge from this?

In my experience advising B2B SaaS companies through FirstMotion, a properly redesigned team can open a meaningful cost and speed advantage within two to three quarters. That’s fast enough to matter well before any broader economic debate about jobs gets resolved.

Should smaller companies worry about this as much as large enterprises?

Arguably more so. A smaller team has less slack to absorb inefficiency, so the leverage from redesigning roles around AI tends to be proportionally larger for a leaner organisation than for a large enterprise with more resources to spare.

Does VAQA help with this kind of operational redesign?

Yes, this sits squarely under my Growth and Operational Efficiency advisory pillars at VAQA, often alongside Go-to-Market work since the two are usually connected. I work directly with founders and leadership teams on the practical task level changes, not just high level strategy decks.

Isn’t it risky to redesign roles before the technology matures further?

The tools will keep improving regardless of when you start, so waiting doesn’t remove that uncertainty, it just delays the benefit. I’d rather build an organisation that adapts its role design every quarter than one that waits for a finished, mature version of AI that may never arrive.

Tom Batting is a Forbes 30 Under 30 entrepreneur, founder of Obby and Baluu, and founder of FirstMotion. He advises founders and leadership teams through VAQA on Board Advisory, Growth, Go-to-Market, AI Implementation, Operational Efficiency, and Finance and Fundraising.